OpenZhenZhen AI 提示词画廊 · gpt-image-2.5 · image

凌晨两点的秘密俯视视角

Secret Overhead View at 2 AM

A flexible observational template for a weekend-night 2 AM overhead scene, using a fictional adult and a concrete quiet activity so the brief is directly executable.

凌晨两点的秘密俯视视角 — AI image creation
凌晨两点的秘密俯视视角

提示词 / Prompt

中文提示词

制作一幅电影感的高处正俯视场景,时间必须是周末夜晚凌晨 2 AM,镜头从上方垂直向下,像秘密观察一样看看一位虚构成年人正在做什么。默认设定为:一名普通成年人物坐在公寓窗边地板上,在柔和灯光下安静整理打印照片和笔记本,旁边放着正在变凉的茶杯。表现清楚的房间布局、散落照片、地毯纹理、窗光、时钟微光和长阴影,使俯视角度具有观察叙事,而不是摆拍肖像。人物必须完全虚构且不可识别,穿着遮蔽充分、舒适的居家睡衣。使用深蓝夜间氛围、一盏小型温暖实用灯、真实的编辑摄影、克制颗粒和亲密故事感。不表现偷窥式侵入,不出现标志、水印或额外文字。

English prompt

Create a cinematic overhead-view scene at exactly 2 AM on a weekend night, looking straight down from above to observe what one fictional adult is doing. Use the default: a generic adult sitting on the floor beside a softly lit apartment window, quietly sorting printed photographs and a notebook while a mug of tea cools nearby. Show the room layout, scattered photographs, rug texture, window light, clock glow, and long shadows so the overhead angle feels like a secret observational peek rather than a posed portrait. Keep the subject fully fictional and non-identifiable, in opaque comfortable nightwear. Use deep blue night ambience, a small warm practical lamp, realistic editorial photography, restrained grain, and intimate storytelling. No voyeuristic intrusion, logos, watermarks, or extra text.

查看 API 模型

探索更多图片和视频提示词 · 生成图片 · 生成视频 · 在 AI 画布中创作

  • 1girl

    gpt image2.5生成

  • 【超必杀】圣光白米大肥鱼

    整体基调:东方史诗神话灾难大片质感,热血动漫爆发感融合好莱坞写实CG,UE5电影渲染,HDR光追,真实物理,强烈金色侧逆光,体积沙尘,电影景深,动态运动模糊,节奏干脆迅猛。 图1锁定角色:蓝渐变长卷发,头顶呆毛,鲸鲨鱼耳,鲸鲨长尾,女仆裙装,白色围裙,围裙印有小鲸鱼图案,Q版比例。 图2锁定巨型黑黄条纹甲壳虫王,巨口獠牙、尖刺前爪、虫群头领气质。 图3只用于冲击帧视觉语言【严格遵循Impact Frame规则】:冲击帧不是连续动画,不是撕裂米粒浪潮喷涌过程,也不是爆炸过程,而是**米粒旋风接触虫王躯干的命中顶点瞬间,强行打断正常运动、插入极短静态定格画面的视觉卡点**。冲击帧仅0.2秒,帧内无

  • 法式风情地标美食海报

    A premium, clean and elegant poster concept built from the source’s broad request. Eiffel Tower, Arc de Triomphe, ratatouille and Provençal bouillabaisse are curator-selected illustrative examples, not details stated by the source; they can be replaced in a future variant. The source example image i

  • 梦回江户浮世绘古典人物像

    A reference-led Edo-period ukiyo-e portrait transformation that preserves the supplied person’s facial identity while restyling hair, clothing and accessories in traditional Japanese fashion. The current ref-10a-v1 is the generic person preview for this workflow, not a claim about any external named

  • 盛世长安万国来朝图

    A text-driven historical reconstruction of the source scene: an aerial view of Kaiyuan-era Chang’an, Emperor Xuanzong and Yang Guifei looking out from Hanyuan Hall, and a distant tribute procession. The configured generic portrait is used only as a face-identity anchor for Yang Guifei in this genera

  • 威震天高定西装时尚大片

    A text-driven fashion-cover portrayal of Megatron, retaining the source's silver-gray mecha, haute-couture suit and cyberpunk neon setting; it does not claim an exact adaptation-specific design. / 依据文字创作的威震天时尚封面,保留银灰机甲、高定西装与赛博朋克霓虹场景;不声称复现某一改编版本的精确造型。